Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
| Publication date | 2008 |
|---|---|
| Host publication | PM2HW2N'08: Proceedings of The Third ACM International Workshop on Performance Monitoring, Measurement, and Evaluation of Heterogeneous Wireless and Wired Networks |
| Editors | Special Interest Group in Simulation |
| Place of Publication | New York |
| Publisher | Association for Computing Machinery (ACM) |
| Pages | 1-8 |
| Number of pages | 8 |
| ISBN (print) | 9781605582399, 1605582395 |
| <mark>Original language</mark> | English |
| Event | 3rd ACM Interntional Workshop on Performance Monitoring, Measurement, and Evaluation of Heterogeneous Wireless and Wired Networks - Vancouver, Canada Duration: 31/10/2008 → … |
| Conference | 3rd ACM Interntional Workshop on Performance Monitoring, Measurement, and Evaluation of Heterogeneous Wireless and Wired Networks |
|---|---|
| Country/Territory | Canada |
| Period | 31/10/08 → … |
| Conference | 3rd ACM Interntional Workshop on Performance Monitoring, Measurement, and Evaluation of Heterogeneous Wireless and Wired Networks |
|---|---|
| Country/Territory | Canada |
| Period | 31/10/08 → … |
In this paper, we propose an efficient method to estimate the end-to-end Available Bandwidth (AB) of a network path. This method employs multi-rate (MR) probe packet sequences with Kalman filtering. Our proposed method is an extension of the Bandwidth Available in Real Time (BART) method, so that the probe packet sequences are injected into the network path of interest. Using this technique, the probing rate within each probing sequence is varied. We show that by a marginal addition to the computation complexity compared to the conventional BART technique, the proposed method converges faster than that of BART and achieves a more accurate estimation. In addition, this method is more robust against inappropriate initial value of Kalman filter than the conventional BART method. Furthermore, we obtain the estimation error of MR-BART based on the parameters of the probe packet sequences.